Retinal Fundus Disorder Detection

Overview

This project implements a retinal fundus image classification system using three approaches; naive approach, traditional machine learning and deep learning models. The goal is to analyze and classify retinal images to identify potential disorders such as Diabetic Retinopathy, Glaucoma, AMD, Cataracts, etc..

Approaches

Naïve Rule-Based Classifier

• Uses Edge Density, Brightness, and Optic Disc detection for rule-based classification.

Traditional Machine Learning (SVM Model)

• Uses Local Binary Pattern, Color Histograms (LAB space), and Vessel Magnitude as features; using SVM classifier with PCA for dimensionality reduction.

Deep Learning (EfficientNet-B4 CNN)

• A fine-tuned EfficientNet-B4 model with custom classifier layers.

User Interface

• Streamlit Web App: UI to classify uploaded images and display prediction probabilities for all three approaches.

Setup

./setup.sh

This script takes care of setting up your virtual environment if it does not already exist, activating it, installing requirements, pulling the dataset (if not already present in the data directory), pre-processing the data for the traditional model (feature extraction), and traditional model training.

At any point that a matplotlib graph is opened, please close it to continue the script execution.

Deep Learning Model Training

Assuming your virtual environment is setup and activated, and that the requirements are installed from running setup.sh, you can then train the deep learning model.

python scripts/deep/model.py

Note: you should run this script from a device with a GPU/TPU. The script supports CUDA and MPS. This script isn't included in setup.sh because it takes hours to run.

Running the Streamlit application locally

Assuming your virtual environment is setup and activated, and that the requirements are installed from running setup.sh, you can then run the following to startup a local instance of the Streamlit application.

python main.py

Dataset & License

This repository uses the Retinal Fundus Images dataset from Kaggle, licensed under CC-BY-NC-SA 4.0.

Canva Presentation

View our Canva presentation HERE.

Streamlit Application

Access our Streamlit application HERE.